BALM-Benchmark is a curated collection of datasets designed to advance machine learning and deep learning model research for protein-ligand binding affinity prediction. This benchmark consolidates several key datasets including BindingDB, LP-PDBBind, and specific protein-ligand systems like USP7, MPro, SYK, HIF2A, and MCL1, each chosen for its distinct data characteristics and evaluation.
This dataset collection has been refined and standardized… See the full description on the dataset page:
https://huggingface.co/datasets/BALM/BALM-benchmark.
